changing lifters...

[isham-research.freeserve.co.uk at pop.pol.net.uk]

> Phil:  I'm  now looking at a cross section through the design of the
> hydraulic lifters used in the Audi I5 engine.  I suggest you find one and do
> the same.

The best diagram Audi published is the one in the V8Q Introductory,
where they contrast the designs used in the older cars with the new
design introduced with the V8Q.  It's a few years since I looked at
it in detail - it's basically a very simple device.

> In my opinion, there is no way the lifter, once worn, will hold  oil
> pressure inside it if the camshaft, if once stopped, has the camshaft lobe
> pressed against the lifter which has to resist the valve spring  tension for
> several hours.  It _will_ leak down and create noise on start up.  A
> significantly worn lifter will leak down by gravity alone without the
> additional valve spring pressure.

The lifter is designed to collapse in exactly this way - if it didn't,
it wouldn't work at all.  This is also why there is a note in the
hardware manual for each and every hydraulic lifter engine stating that
the engine must not be started for 30 minutes after new lifters have
been fitted.

(Yes, I've done this - I once got an MB back together in 22 minutes
 and had to wait.  Takes effort, though, and a power screwdriver to put
 the cam cover nuts back on.)

If you ever get a stuck oil pressure relief valve leading to significant
oil overpressure, one symptom is no compression - the lifters get
pumped up with oil faster than they can leak it down.

> I agree with Audi/VW.  That's why you get intermittent hydraulic lifter
> noise on startup, not on all the lifters, but the ones that when at rest
> were under valve spring tension, or the ones that have serious leak down
> problems.  In my calcs that would be between 2 to 4 lifters(on a 10V engine)
> depending on where the  camshaft stops, and, for how long is at rest.
> (think about the pressure inside the lifter from the valve spring, its far
> more than the anti drain back valve in the oil filter would hold even if it
> was a fantastic design.  Take one apart and look at it.)

Full leakdown of the lifters under pressure occurs in less than a minute
after the engine stops.  If the engine's hot and/or it has a thin
synthetic oil - a lot less.

> I'm sure that's why Audi says the lifter noise is no big deal if it goes
> away shortly after startup.(2 minutes is generous, mine (WU engine @ 100,000
> miles)went away after about 2-5 seconds)   No matter what oil/oil filter I
> used, no matter what the oil level,  in the past, with worn lifters (100,000
> miles plus)  made noise on startup but functioned perfectly shortly
> thereafter.

The official specification for the MC/MB engines is:

a) Idle the engine until the radiator fan cuts in.

b) Raise rpm to 2500 for 2 minutes

c) Drop to idle and listen.

If one or more tappets rattle, stop the engine and remove the cam
cover.  Turn the cam to lift the lobes off any suspect tappet and test
for free travel before the valve starts to open.  More than 0.1mm is not
acceptable.

> Like anything, I'm sure there is a limit to the wear, where the lifter will
> no longer hold sufficient oil pressure during operation, makes continuous
> noise, and needs to be replaced.  On start up, I'm not sure that is the time
> to make the decision, unless you hate being embarrassed by the noise.

The most common thing you hear is marginal lifters.  Each lifter is fed
with oil under pressure from a jet in the side of its guide.  The oil
enters the lifter through another hole - sometimes directly, sometimes
via a groove around the side of the lifter.  When a lifter is starting
to go bad, the direct feed (when it occurs) is enough to keep it up to
performance, but the indirect feed (from around the other side of the
lifter) isn't.  So the lifter sometimes taps and sometimes doesn't.

The lifters (and you have to look VERY carefully) aren't directly under
the cam lobes - they're offset to one side by a fraction of a millimetre.
The best way to see this is under a low-power stereoscopic microscope -
you'll see that the 'radial' wear marks aren't actually radial - they
don't quite meet in the middle.  The effect of this is to rotate the
lifter while the engine is running, thereby distributing wear around
the whole top of the lifter.

A side-effect, though, is to alternately present the lifter with direct
and indirect oil feeds - jet to hole, and jet to back of lifter -
cyclically.  It seems to take between 10 and 20 seconds for a lifter to
rotate at idle in a hot engine - this is the cause of cyclic lifter
tapping - now you hear it, now you don't.

(Cyclic tapping on a solid lifter engine is an entirely different
 phenomenon, usually caused by a particle of dirt under one edge of
 a shim.)

Now that we know who makes the things, we can get new 10V lifters for
the equivalent of $9 each.  It's just not worth the diagnosis effort
to find out which one's gone.

Why filling oil to the second row of dots below 'Max' works so often is
beyond me, but it does.  I have a theory that it primes the oil pump,
or prevents reverse siphonage out of the oil pump when it stops.

--
 Phil Payne
 http://www.isham-research.freeserve.co.uk/quattro
 Phone +44 7785 302803   Fax: +44 7785 309674